cocoindex
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external, potentially untrusted sources such as local files, Amazon S3 buckets, and Postgres databases.
- Ingestion points: The skill uses
cocoindex.sources.LocalFile,cocoindex.sources.AmazonS3, andcocoindex.sources.Postgresto load external content into the agent's processing context, as documented inSKILL.mdandreferences/flow_patterns.md. - Boundary markers: The provided instructions do not mandate the use of specific delimiters or system instructions to prevent the LLM from executing commands potentially embedded within the ingested data.
- Capability inventory: The skill possesses the capability to perform LLM-based structured extraction via
cocoindex.functions.ExtractByLlmand export results to various database targets includingPostgres,Qdrant,LanceDB, andNeo4j. - Sanitization: There is no evidence of mandatory sanitization, filtering, or validation of external content before it is interpolated into LLM prompts for transformation and extraction.
Audit Metadata